How to Run Subgroup Analysis
Use respondent-level variables to inspect whether effects differ across groups.
The Short Version
Use respondent-level variables to inspect whether effects differ across groups. A practical way to begin is to collect subgroup variables before treatment, then open Results. Finish by making sure you treat subgroup findings cautiously unless preregistered. The full walkthrough below explains each part in order.
Is This Guide for You?
Researchers testing moderation by party, ideology, gender, education, or other pretreatment variables.
Follow These Steps
Work through these 5 steps at your own pace. The wording is intentionally practical, and you can return to the checklist at the end when you are ready to review your work.
- Collect subgroup variables before treatment.
- Open Results.
- Choose subgroup fields in the analysis controls.
- Export model-ready data for full interaction models.
- Treat subgroup findings cautiously unless preregistered.
What This Helps You Accomplish
Subgroup analysis can be useful, but it multiplies comparisons quickly. Clear preregistration and exports help keep it defensible.
A Quick Confidence Check
- Use pretreatment moderators.
- Watch small cell sizes.
- Label exploratory tests clearly.
Common Questions
Can I use this guide if I am new to this?
Yes. Researchers testing moderation by party, ideology, gender, education, or other pretreatment variables. Follow the steps in order, start with a small test, and use the final checklist before you field or report anything important.
What is the simplest way to get started?
Begin by collect subgroup variables before treatment. Next, open Results. You do not need to perfect every setting before running a small preview or pilot.
How do I know when I am ready?
Use the confidence check above. In particular: Use pretreatment moderators. Watch small cell sizes. Label exploratory tests clearly. When the decision is consequential, keep your study documentation and ask a qualified colleague or methods reviewer to examine the design as well.
Related Guides
Research artifact: sample data, codebook, and scriptsResearch Pathways
Product names and fielding platforms mentioned in this guide belong to their respective owners. Conjoint Survey is an independent academic research tool and is not affiliated with, sponsored by, or endorsed by those services.